Enabling Data-Driven Decisions in a Big Data World: My Co-op at Activision (Demonware)

Enabling Data-Driven Decisions in a Big Data World: My Co-op at Activision (Demonware)

By: David Quong
  2905 reads

In today’s digital age, data has acquired a prized status among companies and is used to drive the most important of business decisions. My first experience and taste of the “Big Data” came with my co-op at Activision. Our team is located at Demonware’s Vancouver office, which is a subsidiary of Activision­Blizzard. It is composed of data engineers and architects, and operation engineers who are hard at work to develop, maintain and operate the data services platform for some of Activision’s largest video game titles, including but not limited to their widely acclaimed ‘Call of Duty’ series.

 When I was an incoming co-­op student in the data services team, I had no idea what to expect. As an avid gamer, and one who is all too familiar with the games that are developed by Activision, I had anticipated that the volume of data that this team would be working with would be quite high. However, with their most recently released titles (i.e., Call of Duty WW2), which have hundreds of thousands to millions of concurrent users at any given time throughout the day, the amount of data that our platforms provide and manage is insurmountably huge.

 During my first few weeks, I found myself learning about Activision's Extract­Transform­Load (ETL) data pipeline, which handled the archiving and transformation of raw data from the player’s console or PC to a format that was useful and accessible to Activision’s data analysts. Here, I began to actually see many of the technologies that I had touched upon in my classes at SFU, in use. It was interesting to see that the theoretical concepts that I had learned about could be applied practically. This included the use of some of the latest big data technologies and services, including Apache SparkHive, and more.

During my co-­op term, I worked on developing and integrating automated schema inference for the platform’s various data-­processing batch jobs. This feature would reduce the amount of work required by the data architects and engineers when adding or modifying the schema of our exported database tables. What was noteworthy about working at Activision (and Demonware) however, was their unique and quirky company culture. I am sure, all those who are interested in video game development have heard numerous horror stories about meeting hard deadlines, and the implied overtime and harsh working conditions associated with it. Instead, I was pleasantly surprised to find the contrary to be true here in Activision’s data services team and also throughout Demonware. It was apparent that employees’ integration into the organization and them maintaining a work-life balance was a core value that was considered important by the team and the company. Employees are provided with flexible hours, an excellent amount of vacation time (even for co-­ops), and plenty of leisurely events for everyone to participate in.

If you are currently a student seeking a co-­op or are an upcoming graduate soon to be entering the job market, I would highly recommend you consider applying to Demonware. If you are looking to get into the field of big data or video game (services), then I believe Activision’s data services team will suit your interests and will provide excellent exposure to both industries. The data services team will without a doubt provide opportunities to work with new and exciting technologies and tackle problems of a scale unseen elsewhere. Lastly, for students who are looking to gain prerequisite knowledge and skills to excel in this type of position, I recommend the following classes - CMPT 353 (Introduction to Data Science), CMPT 354 (Database I), and CMPT 474 (Web Systems Architecture).

As an upcoming graduate myself, I plan on using the skills and experiences that I have gained from Activision to seek employment in the field of software development with a specialization in designing reliable and scalable systems.  

Beyond the Article 

Posted on July 21, 2018